Perioperative metabolic alkalemia is more frequent than metabolic acidemia in major elective abdominal surgery.

OBJECTIVE

To investigate the incidence, type and etiology of perioperative metabolic disturbances associated with major abdominal surgery. We hypothesized that metabolic alkalemia is more frequent than metabolic acidemia.

METHODS

This was a prospective, observational study, performed in a university-affiliated hospital. 98 consecutive patients undergoing major abdominal surgery were included in the study. Patients were observed by serial vital signs and laboratory measurements during the preoperative, intraoperative, PACU and the first three postoperative day periods. Central venous pressure, systolic pressure variation, fluid input, urine output, temper- ature, electrolytes, and acid-base variables were recorded. The primary endpoint of the study was the incidence of metabolic alkalemia or acidemia. Metabolic alkalemia was defined as pH >7.45 and BE >+3. Metabolic acidemia was defined as pH <7.35 and BE <-3. Continuous variables were described as mean ± standard deviation. Distributions of continuous variables was assessed for normalty using the Kolmogorov-Smirnov test (cut off at P = 0.01). The frequency of metabolic acidemia or alkalemia was compared across time points using Cochran's Q test and between time points using the binomial distribution.

RESULTS

Metabolic acidemia occurred only intraoperatively and in the PACU. Subjects with metabolic acidemia were older, (74 ± 9 yr. vs. 66 ± 12, P = 0.01). Intraoperative body temperature was inversely associated with PACU lactate (P = 0.035). Blood loss >500 mL was more frequent in acidemic patients (42% vs. 19%, P = 0.033). More patients with hyperphosphatemia had acidemia than subjects without hyperphosphatemia (39% vs. 17%, P = 0.019). Metabolic alkalemia occurred more frequently than metabolic acidemia (49% vs. 23%, P < 0.0001) and was correlated with hypochloremia. The incidence of metabolic alkalemia decreased from baseline to intraoperative and PACU periods (13% vs. 3%, P = 0.003) and increased from the PACU to the three postoperative days (3% vs. 45%, P = 0.007).

CONCLUSIONS

Metabolic alkalemia occurred more frequently than metabolic acidemia and occurred mainly preoperatively and postoperatively, while acidemia occurred mainly during surgery and in the PACU.